There’s no denying the buzz around Artificial Intelligence (AI) in IT Service Management (ITSM). According to OTRS Group’s 2024 Spotlight survey, interest in AI for ITSM has nearly tripled in the past year. In 2023, just 24% of U.S. IT teams planned to introduce AI into their workflows. This year, that figure has jumped to 67%.
But with great hype comes great expectation and pressure.
Interestingly, the industry remains torn. Half of IT leaders see more promise in AI, the other half lean toward traditional automation. Regardless of which side of the fence you’re on, one thing is clear, nearly everyone agrees these technologies are now essential for future ITSM success.
What’s Holding Teams Back?
Despite all the interest, many teams are struggling to move beyond the exploratory phase.
Regulatory pressures have taken the lead as the biggest blocker (29%), displacing last year’s top concerns of budget and staffing. Compliance, governance, and emerging AI policy frameworks are now front and centre for IT leaders.
But that’s only part of the story.
Software limitations and a growing need to train staff up from 25% last year to 58% this year highlight a broader readiness issue. Many IT teams already own AI or automation tools, yet almost half admit they’re still figuring out how to use them effectively.
A big reason? Data quality.
AI in ITSM depends on structured, reliable data yet many environments are still grappling with outdated CMDBs, inconsistent ticket records, and siloed service information. Without clean, trusted data, AI models struggle to deliver meaningful outcomes, and automation efforts stall.
For example, some teams have rolled out AI-driven chat-bots, only to find they repeatedly surface irrelevant knowledge articles or fail to triage incidents correctly because ticket categories were misused or poorly defined. In other cases, attempts to automate change workflows have backfired due to missing or conflicting service relationships in the CMDB, creating more friction instead of reducing it.
There are real-world cases that back this up. A small IT service provider in the nautical tourism sector failed to get value from AI-powered incident classification due to poor ticket data and inconsistent record keeping. Even large enterprises aren’t immune, a national healthcare provider faced over 55,000 data issues in its CMDB before AI-based remediation led to gains in efficiency and cost reduction. And Microsoft and Endava’s support ticket classification initiative hinged entirely on cleaning and structuring historical ticket data before deploying machine learning.
So what we’re seeing is a maturity gap, not in interest or investment, but in enablement. It’s not just about having the tools, it’s about making sure the people, processes, and data are ready to support them.
The Payoff for Those Who Get It Right
Despite these hurdles, the organisations that manage to get it right are seeing tangible gains. According to OTRS, AI implementation has led to.
- 83% improvement in service availability
- 80% reduction in resolution times
- 74% increase in knowledge article creation
- 74% boost in request completion rates
Goldman Sachs offers a striking example of what’s possible when AI is integrated thoughtfully. By deploying Microsoft GitHub Copilot across its development teams, Goldman reported a 20% productivity gain, reducing project timelines from months to weeks. Critically, they took steps to centralise AI efforts under strict governance, ensuring that security and compliance were baked into their strategy from the outset.
This approach reflects a broader shift. AI in ITSM isn’t just about speed, it’s about trust, structure, and strategic enablement.
The Bottom Line
AI has a clear role to play in improving ITSM, but success depends on more than installing new tools. It requires a readiness to rework data foundations, upskill teams, and embrace governance practices that ensure responsible, value-driven use.
Get the foundation right, and the technology will follow.